Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    Code/math matrices

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    Residential
    -0.09
    Referral
    -0.09
    /story
    -0.09
     psychological
    -0.09
     celebrates
    -0.09
     narrator
    -0.09
    Dental
    -0.09
     psychos
    -0.09
    Severity
    -0.09
     referral
    -0.09
    POSITIVE LOGITS
     matrices
    0.18
     matrix
    0.17
    Matrices
    0.17
    matrix
    0.14
    矩
    0.14
    	Matrix
    0.14
    _MATRIX
    0.14
    Transpose
    0.13
    Matrix
    0.13
    	matrix
    0.13
    Act Density 0.031%

    No Known Activations